With a clear idea about your research problem and a sound knowledge of the related literature, your next step should be to make a preliminary choice of methodology. In doing so, you should distinguish between three related concepts: research perspectives, research types, and research methods. This chapter will provide only an overview of these concepts and suggest how to make a preliminary choice of methodology. You will need to consult with a faculty member who specializes in research design and methods to be sure that your methodology is sound. Chapter 9 provides additional detail about explicating the methodology for the proposal; Chapter 16, for the dissertation.
A research perspective, as used here, is a general view and use of research approaches and methods. There are two major perspectives: quantitative and qualitative. The quantitative perspective derives from a positivist epistemology, which holds that there is an objective reality that can be expressed numerically. As a consequence, the quantitative perspective emphasizes studies that are experimental in nature, emphasize measurement, and search for relationships. If a study uses language such as the following, it probably has used a quantitative perspective: variable, controls, validity, reliability, hypothesis, statistically significant.
On the other hand, a qualitative perspective emphasizes a phenomenological view in which reality inheres in the perceptions of individuals. Studies deriving from this perspective focus on meaning and understanding and take place in naturally occurring situations (McMillan, 2012). If a study uses language such as the following, it probably has used a qualitative perspective: naturalistic, field study, case study, context, situational, constructivism, meaning, multiple realities.
Although some researchers seem chiefly concerned with the differences between the two approaches, Creswell (2009) and Morgan (1997) explain how the two perspectives can be combined. Four general ways of combining the two were identified by them, based upon two factors: which one is primary and which secondary, and which one is used first and which second.
Project-based dissertations are an emerging trend as it permits students to use current practice as the basis for research; the project is usually designed to evaluate the practice—research—for effectiveness. For this type of study, some universities permit students to work together as groups to complete the research; however, the framework for the research must be grounded by research theory. Typically, the first part of the study is spent observing the practice, while the second component is to conduct a detailed evaluation of the practice and its effectiveness in reaching the intended outcomes. Be cautious when considering this type of dissertation, because many graduate faculties may not be familiar with this type of study, and those faculties may not think that the final product is a dissertation.
The term research type is used here to identify the general research approach. Although authorities in the field seem to differ as to how the types of research are classified, the following approaches, which are most often used in educational research, represent some of the options available to you as a researcher. To simplify the discussion, they are divided into whether they tend to use a quantitative or a qualitative perspective, although there is much overlapping in many of the types.
The following types of research are primarily quantitative in nature.
Experimental research uses methods originally applied in the physical and biological sciences. In most experiments, the following procedures are used: a sample of subjects is selected, they are assigned randomly to experimental and control groups, and a treatment is administered to the experimental group only. The two groups are then evaluated on the basis of the dependent variable and the consequence of the independent variable. The latter is the presumed cause of the dependent variable.
Here is an example of an experimental design. The researcher randomly assigns 200 fourth graders to 10 classrooms. Five of those classes are randomly selected. They are given a new program on ethics; the other five classes are given no program at all. At the end of the school year, both groups are administered a measure that assesses their attitudes about and knowledge of ethics. Statistical techniques are used to analyze the results.
A quasi-experimental design is one that follows the general procedures of experimental research, without the use of a control group or without random assignment, since random assignment or the use of control groups is often not feasible in educational settings. In a quasi-experimental study of ethics, the researcher would not use random assignment but could use one of several strategies to compensate for its absence. The researcher might, for example, administer a pretest to all students, provide the treatment to half of the students, administer a posttest, and use a statistical procedure known as the analysis of covariance to determine if the treatment made a difference.
Causal-comparative studies are designed to determine the possible causes of a phenomenon. Sometimes these studies are called ex post facto research since the causes are usually studied after they have had an effect upon another variable. For example, a study which examines the effects of grade retention uses a causal-comparative design by studying the effects of not being promoted when comparing the later performance of those who were retained and those who were promoted.
Correlation studies are designed to analyze the relationships between two or more variables, ordinarily through the use of correlation coefficients. You might, for example, study the type of licensure held by teachers and student achievement to determine if there are any statistically significant relationships between these two factors. Keep in mind the caution that correlation is not causation. Correlational studies may show a direct relationship between two factors but cannot prove causation.
As the term implies, the purpose of descriptive research is to describe a phenomenon. Although some professors deprecate descriptive studies, these studies can be especially valuable as one of the early stages in a research project. Descriptive studies report frequencies, averages, and percentages. For example, you might study the attitudes of teachers in public schools toward the superintendent and the local governing board. You would not draw any conclusions about relationships, but only report frequencies, averages, and percentages.
Evaluation research makes judgments about the merit or worth of educational programs, products, and organizations. It is typically undertaken to aid administrators in making professional decisions. Evaluation studies are usually described as either formative or summative. Formative studies are made while a new program or product is being developed; summative studies, when it has been completed. You might do an evaluation of a testing program required by the No Child Left Behind legislation, performing both a formative and a summative assessment.
The following types of research tend to take a qualitative perspective.
A very useful definition of the case study is that provided by Yin (2009): “A case study is an empirical inquiry that investigates a contemporary phenomenon in depth and within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident” (p. 18). Although, as Yin and others state, case studies often use quantitative measures, they more often tend to take a qualitative perspective, concerned with exploring, describing, and explaining a phenomenon. You might undertake a case study of a new consolidated high school in its first year of operation.
Ethnographic research is a special type of case study research. It is distinguished from other types of case tudies because it uses the theories and methods of anthropology to study the culture of schools and classrooms. For example, you might do an ethnographic study of the educational background and employment experiences of female school administrators to use in curriculum planning for preparing female administrators.
Most action research documents how an educational problem was identified, understood, and solved by practitioners. For example, you might document how you and your colleagues resolved the problem of student “texting” during class at the elementary, middle, and secondary school levels.
There are, of course, other methods employed in educational research, such as philosophical research, historical research, and legal research. Each of these has its own characteristics and uses special methods.
Research methods, as the term is used here, are the specific techniques used to collect data with respect to the research problem. Five methods are typically used in educational research.
Exhibit 5.1 Relationship of Types and Methods
CODE: P, primary method used; A, additional method that may be used.
As Exhibit 5.1 indicates on the preceding page, the types of research explained above make more use of certain methods than they do of others, although there is much variation in the relationship of types and methods.
The process explained here assumes that in developing the prospectus you will make only a preliminary choice that may be modified through discussion with your committee. When making a preliminary choice regarding the research methodology, the type of study, and data collection methods, weigh the following factors.
As you consider the various research methodologies and designs, you need to consider the type of data to be collected and the process and procedures that will be used to analyze the collected data. Data analysis software is available for both qualitative and quantitative studies. Please keep in mind that any collected data must be in computer readable form before any analyses can occur.
Quantitative data are much easier than qualitative data to analyze as the analysis or analyses can be completed quickly using statistical data analysis software. The challenging aspect of quantitative data is preparing the data for analysis as the collected data will need to be “computerized” in some form. Previously, a data entry operator would key the data; however, today’s technology allows the scanning of such data into computer readable form. Examples of quantitative microcomputer data analysis packages are Statistical Analysis Software (SAS), Statistical Package for the Social Sciences (SPSS), SysStat, and MiniTab.
Qualitative data are more challenging to analyze as the data are not numerical but text, or words. As a part of the analysis process, the words are to be coded in a format that is computer recognizable. For the data to be correctly coded, the same person must code all the data. If it is not feasible for one person to code the data, then a review process needs to be implemented. The review process, which is a verification of the coding, requires another individual or individuals to review what has been coded to ascertain that all individuals involved perceive and interpret the data in the same manner. NVivo is the leading qualitative data analysis software, and many universities have the software available for faculty and graduate students.
Dissertating students should take advantage of technology when analyzing data. Regardless of the type of research conducted, data analysis software should be used to streamline the analysis process.